Buckets:

rtrm's picture
download
raw
4.11 kB
import{s as Q,n as I,o as K}from"../chunks/scheduler.37c15a92.js";import{S as V,i as W,g as l,s,r as z,A as X,h as i,f as a,c as n,j as G,u as B,x as v,k as J,y as Z,a as o,v as O,d as j,t as A,w as S}from"../chunks/index.2bf4358c.js";import{Y as ee}from"../chunks/Youtube.1e50a667.js";import{C as te}from"../chunks/CourseFloatingBanner.6add7356.js";import{H as ae,E as oe}from"../chunks/getInferenceSnippets.1837c472.js";function se(U){let r,b,_,w,m,C,d,E,c,T,f,k="Os modelos de encoder (codificadores) usam apenas o encoder de um modelo Transformer. Em cada estágio, as camadas de atenção podem acessar todas as palavras da frase inicial. Esses modelos geralmente são caracterizados como tendo atenção “bidirecional” e são frequentemente chamados de <em>modelos de codificação automática</em>.",P,p,N="O pré-treinamento desses modelos geralmente gira em torno de corromper de alguma forma uma determinada frase (por exemplo, mascarando palavras aleatórias nela) e encarregando o modelo de encontrar ou reconstruir a frase inicial.",M,u,Y="Os modelos de codificador são mais adequados para tarefas que exigem uma compreensão da sentença completa, como classificação de sentença, reconhecimento de entidade nomeada (e, mais geralmente, classificação de palavras) e resposta extrativa de perguntas.",L,g,D="Os representantes desta família de modelos incluem:",R,h,F='<li><a href="https://huggingface.co/transformers/model_doc/albert.html" rel="nofollow">ALBERT</a></li> <li><a href="https://huggingface.co/transformers/model_doc/bert.html" rel="nofollow">BERT</a></li> <li><a href="https://huggingface.co/transformers/model_doc/distilbert.html" rel="nofollow">DistilBERT</a></li> <li><a href="https://huggingface.co/transformers/model_doc/electra.html" rel="nofollow">ELECTRA</a></li> <li><a href="https://huggingface.co/transformers/model_doc/roberta.html" rel="nofollow">RoBERTa</a></li>',q,$,H,x,y;return m=new ae({props:{title:"Modelos codificadores",local:"modelos-codificadores",headingTag:"h1"}}),d=new te({props:{chapter:1,classNames:"absolute z-10 right-0 top-0"}}),c=new ee({props:{id:"MUqNwgPjJvQ"}}),$=new oe({props:{source:"https://github.com/huggingface/course/blob/main/chapters/pt/chapter1/5.mdx"}}),{c(){r=l("meta"),b=s(),_=l("p"),w=s(),z(m.$$.fragment),C=s(),z(d.$$.fragment),E=s(),z(c.$$.fragment),T=s(),f=l("p"),f.innerHTML=k,P=s(),p=l("p"),p.textContent=N,M=s(),u=l("p"),u.textContent=Y,L=s(),g=l("p"),g.textContent=D,R=s(),h=l("ul"),h.innerHTML=F,q=s(),z($.$$.fragment),H=s(),x=l("p"),this.h()},l(e){const t=X("svelte-u9bgzb",document.head);r=i(t,"META",{name:!0,content:!0}),t.forEach(a),b=n(e),_=i(e,"P",{}),G(_).forEach(a),w=n(e),B(m.$$.fragment,e),C=n(e),B(d.$$.fragment,e),E=n(e),B(c.$$.fragment,e),T=n(e),f=i(e,"P",{"data-svelte-h":!0}),v(f)!=="svelte-qj02cj"&&(f.innerHTML=k),P=n(e),p=i(e,"P",{"data-svelte-h":!0}),v(p)!=="svelte-9pg3ea"&&(p.textContent=N),M=n(e),u=i(e,"P",{"data-svelte-h":!0}),v(u)!=="svelte-9o6rgt"&&(u.textContent=Y),L=n(e),g=i(e,"P",{"data-svelte-h":!0}),v(g)!=="svelte-1x0lt16"&&(g.textContent=D),R=n(e),h=i(e,"UL",{"data-svelte-h":!0}),v(h)!=="svelte-18kzzol"&&(h.innerHTML=F),q=n(e),B($.$$.fragment,e),H=n(e),x=i(e,"P",{}),G(x).forEach(a),this.h()},h(){J(r,"name","hf:doc:metadata"),J(r,"content",ne)},m(e,t){Z(document.head,r),o(e,b,t),o(e,_,t),o(e,w,t),O(m,e,t),o(e,C,t),O(d,e,t),o(e,E,t),O(c,e,t),o(e,T,t),o(e,f,t),o(e,P,t),o(e,p,t),o(e,M,t),o(e,u,t),o(e,L,t),o(e,g,t),o(e,R,t),o(e,h,t),o(e,q,t),O($,e,t),o(e,H,t),o(e,x,t),y=!0},p:I,i(e){y||(j(m.$$.fragment,e),j(d.$$.fragment,e),j(c.$$.fragment,e),j($.$$.fragment,e),y=!0)},o(e){A(m.$$.fragment,e),A(d.$$.fragment,e),A(c.$$.fragment,e),A($.$$.fragment,e),y=!1},d(e){e&&(a(b),a(_),a(w),a(C),a(E),a(T),a(f),a(P),a(p),a(M),a(u),a(L),a(g),a(R),a(h),a(q),a(H),a(x)),a(r),S(m,e),S(d,e),S(c,e),S($,e)}}}const ne='{"title":"Modelos codificadores","local":"modelos-codificadores","sections":[],"depth":1}';function re(U){return K(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class fe extends V{constructor(r){super(),W(this,r,re,se,Q,{})}}export{fe as component};

Xet Storage Details

Size:
4.11 kB
·
Xet hash:
e128ad6277d64ee516faa1a51dde4b341a9f1b50ff5f2a4048026b7f9374bf85

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.